Abstract Lupus nephritis (LN) is one of the most serious manifestations of systemic lupus erythematosus (SLE) and is associated with significant morbidity and mortality. Current clinical laboratory markers lack sufficient sensitivity and specificity to optimize individual patient care creating a need to develop novel biomarkers for LN. To address this unmet need there has been a significant effort within the lupus research community to identify novel LN biomarkers, but to date none have been qualified for clinical use. We speculate that one of the important causes of these failures is that no single biomarker can account for the heterogeneity of SLE or LN. Under the hypothesis that multiple categories of biomarkers can represent different aspects of risk for LN, we propose to assess biomarkers that are classified into categories of disease susceptibility, systemic inflammation, and kidney compartment/cell injury to build models for risk prediction. Our objective is to develop composite biomarkers for early detection of renal flares, chronic renal damage, and treatment response of LN patients. To accomplish this goal we will use biospecimens from longitudinal observational cohorts and clinical trials that have well-annotated patient samples. The Medical University of South Carolina (MUSC) team therefore proposes to partner with the Ohio State University (OSU) team to combine each group's longitudinal SLE cohorts and create a well-phenotyped patient resource suited to biomarker discovery and large enough to adequately power validation studies. The combined OSU-MUSC cohort will have approximately 500 LN patients who have had rheumatology/nephrology clinic visits and biospecimens collected every two to six months for up to ten years. Using these samples, our collective team has already described several DNA, serum and urine biomarkers that are associated with disease activity, treatment response or organ damage in LN, and has considerable experience in developing biomarker panels for outcomes in LN. We have reviewed the last decade's literature, systematically ranked novel biomarkers, and propose to test the top performing LN biomarkers in this proposal to address whether: (i) Biomarkers associated with active LN can be used to predict impending renal flares in prospective longitudinal cohorts; (ii) Baseline measures of composite panels of urine or serum biomarkers can distinguish between who will and who will not achieve a renal response to one year of conventional therapy with mycophenolate mofetil; (iii) The development of genetic risk profiles of LN patients to identify those who are predisposed to develop chronic kidney damage. We will also compare biomarkers predictive of disease outcomes in LN to those in non-renal SLE, working towards developing LN specific vs general biomarker panels that are indicative of active inflammation and/or damage accrual. Our ultimate goal is to have validated LN predictors of sufficient sensitivity and specificity to be clinically relevant, and that can be easily assayed in most clinical service laboratories to improve patient care.